Invasive bacteria must acquire and assimilate carbon substrates from host tissues in order to replicate and persist in vivo. Metabolic pathways that are essential for bacterial survival during infection are potentially interesting targets for antimicrobial drug development. The glyoxylate cycle, which is required for metabolism of fatty acid substrates, is widespread among prokaryotes but absent in vertebrates. Previously we reported that isocitrate lyase 1 (ICL1) and 2 (ICL2), which catalyze the first unique reaction in the glyoxylate cycle, are jointly required for Mycobacterium tuberculosis (MTB) growth, survival, and virulence in vivo. Recently we discovered that fatty acids are markedly toxic towards ICL-deficient bacteria, even when metabolizable substrates such as glucose are present. These observations suggest that fatty acids are catabolized by MTB during infection and the glyoxylate cycle is required to prevent the accumulation of a toxic metabolic intermediate(s). We propose three Specific Aims to test this hypothesis. In AIM 1 we will delete the MTB glcB gene encoding malate synthase, which catalyzes the second unique reaction in the glyoxylate cycle. We predict that growth, survival, and virulence of the glcB mutant will be attenuated in ex vivo-infected macrophages and in aerosol-infected mice and guinea pigs. We predict that the glcB mutant will be unable to grow on fatty acids in vitro and will be killed by them. In AIM 2 we will delete the MTB fadB1, fadB2, and fadB3 genes encoding paralogs of hydroxyacyl-CoA dehydrogenase, which catalyzes the penultimate step in the fatty acid beta-oxidation cycle. We predict that growth, survival, and virulence of the fadB1-3 triple-knockout mutant will be attenuated in macrophages, mice, and guinea pigs. We predict that the fadB1-3 mutant will be unable to grow on fatty acid substrates in vitro but will not be killed by them, due to the mutant's inability to degrade fatty acids. In AIM 3 we will focus on the mechanism of fatty acid toxicity towards ICL-deficient bacteria. We will identify suppressor mutations that allow ICL-deficient bacteria to grow in the presence of fatty acids. We will use 13C nuclear magnetic resonance spectroscopy (13C-NMR) to profile metabolite pools in wild-type versus ICL-deficient bacteria fed 13C-labeled fatty acids. Our goal is to identify a potentially toxic metabolite(s) derived from fatty acid catabolism that accumulates in the absence of ICL. Lastly, we will use microfluidics and video microscopy to analyze the real-time single-cell response of ICL-deficient bacteria after infusion of fatty acids. The studies proposed in this application will elucidate the role of MTB fatty acid metabolism during infection, and could potentially identify novel targets for drug development.